Women taught me how to do the unskilledwork

1. INTRODUCTION

Since the 1970s the global economy has beenin an era of market regulation and growinglabor market flexibility, in which new technol-ogies, new labor control systems and reformedforms of work organization have transformedpatterns of labor force participation through-out the world. In the process, the turn of thecentury will mark the end of the century of thelaboring man in a literal and real sense, in thatwomen will account for almost as many of the‘‘jobs’’ as men.

This paper is a ‘‘revisit’’ to ideas and datapresented in a paper written in 1988.2Themain hypothesis of that paper was that thechanging character of labor markets aroundthe world had been leading to a rise in femalelabor force participation and a relative if notabsolute fall in men’s employment, as well as a‘‘feminization’’ of many jobs traditionally heldby men.

The term ‘‘feminization’’ was intentionallyambiguous. Perhaps a better term could havebeen used. It was intended, however to capturethe double meaning and the sense of irony that,after generations of efforts to integrate womeninto regular wage labor as equals, the conver-gence that was the essence of the original hy-pothesishasbeentowardthetypeofemployment and labor force participation pat-terns associated with women.

The era of flexi-bility is also an era of more generalizedinsecurity and precariousness, in which manymore men as well as women have been pushedinto precarious forms of labor.

Feminization arises because available em-ployment and labor options tend increasinglyto characterize activities associated, rightly orwrongly, with women and because the patternof employment tends to result in an increasingproportion of women occupying the jobs. Theterm could be decomposed into its constituents.A type of job could be feminized, or men couldfind themselves in feminized positions. Morewomen could find themselves in jobs tradi-tionally taken by men, or certain jobs could bechanged to have characteristics associated withwomen’s historical pattern of labor force par-ticipation. The characteristics include the typeof contract, the form of remuneration, the ex-tent and forms of security provided, and theaccess to skill.

A further diffculty arises from the connota-tions. Most observers think that work patternsthat are intermittent, casual and partial arebad, while those that are stable, continuous andfull are good. If the surrounding conditions areappropriate, however there is nothing intrinsi-cally bad about a pattern of work involvingmultiple statuses, multiple activities and vary-ing intensity of involvement in di?erent formsof work.

Gender outcomes in labor markets do notreflect natural or objective di?erences betweenmen and women, but rather reflect the outcomeof discrimination and disadvantage, and thebehavioural reactions by workers and employ-ers. 

This means that even if the thesis of femi-nization were supported empirically, a reversalof trend could still be possible. That stated, thefollowing does no more than bring the originalhypothesis up to date with a decade more ofdata used in the original paper, bearing in mindall the di?culties of making crossnationalcomparisons.

To reiterate, the contextual developmentsthat have shaped the growing feminization ofthe labor market include: (a) International trade in goods and services hasgrown enormously as a share of national in-comes, as has the share of foreign or multina-tional investment in total investment in mostcountries.(b) Trade and investment have been directed in-creasingly to economies in which labor costshave been relatively low (or where they havebeen expected to be relatively low), putting apremium on the level of wages, nonwage laborcosts and labor productivity.(c) In the postwar era up to the 1970s, trade be-tween countries was predominantly in comple-mentary goods (e.g., primary for nonprimary)or between countries with similar labor rights,and therefore roughly equivalent labor costs(balanced by di?erences between wages andproductivity). From the 1970s onward, partlyas a consequence of actual and incipient indus-trialization of some parts of the developingworld, labor rights in industrialized countriesbecame increasingly perceived as costs of pro-duction to be avoided in the interest of enhanc-ing or maintaining ‘‘national competitiveness.’’(d) In the past few years, there has been a ‘‘tech-nological revolution,’’ based on micro-electron-ics, which inter alia has permitted a wider rangeof technological-managerial options in workingarrangements, which again means that cost con-siderations of alternatives have become moresignificant determinants of allocations and divi-sions of labor. This has affected patterns of em-ployment in industrialized and industrializingeconomies, and the international division of la-bor, accentuating tendencies to allocate towhere labor costs are lowest (which dependson wages, nonwage labor costs, productivityand supporting infrastructure).3There is alsothe possibility that we have been in a phase ofwhat some analysts have described as ‘‘techno-logical stalemate,’’ in which process (cost-cut-ting) innovations predominate over productinnovations.(e) There has been a crystallization of a globaleconomic strategy, under the banner of ‘‘struc-tural adjustment,’’ ‘‘shock therapy’’ and othersupply-side economic policies. This strategyhas been associated with radical changes in la-bor market relations, involving erosion of pro-tective and pro-collective labor regulations,decentralization of wage determination, erosionof employment security and a trend to marketregulation rather than statutory regulation ofthe labor market.(f) There has been an erosion in the legitimacyof the welfare systems of industrialized coun-tries. In the era following WWII, for much ofthe world universal social protection within a‘‘redistributive welfare state’’ was regarded asa long-term development goal and as the basisof well-functioning labor markets. The erosionof that model has been due to many factors, in-cluding the rising costs of achieving social pro-tection in the context of high unemployment,the rejection of Keynesianism and its replace-ment by faith in supply-side economics, bywhich public spending is perceived (or present-ed) as ‘‘crowding out’’ private, productive in-vestment, and a loss of faith among welfarestate defenders in its ability to be redistributive.There has been growing privatization of socialprotection and an individualization of social se-curity, whereby more workers have to dependon their own contributions and entitlements.

These contextual developments have bothshaped the gender division of labor and havebeen influenced by the labor market develop-ments themselves. In particular, they have in-creased the emphasis placed on labor costs.That has led to greater use of alternative formsof employment to the conventional one ofregular, full-time wage labor, which has weak-ened the dualistic segmentation of employmentin which men have been relatively protected‘‘insiders’’.

2. GENDER IMPLICATIONS OF LABORMARKET FLEXIBILITY

Among the labor market implications of thesupply-side, structural adjustment agenda pur-sued around the world in recent years, severalare relevant to our general hypothesis.

First, in industrialized countries in particular,the increasing selectivity or ‘‘targeting’’ of statebenefits has meant fewer people having en-titlements.Thishasboosted‘‘additionalworker’’ e?ects — pushing more women intothe labor market in recessions and inducingmore women to remain in the labor marketbecause of the growth of income insecurity. Thetrend to means tests and tighter conditionalityhas also encouraged the growth of the ‘‘blackeconomy’’ and precarious forms of work, sincethose without entitlements have been obliged todo whatever income-earning work they can.This phenomenon has been strong in industri-alized economies, although it has a?ected manyindustrializing countries as well.

Second, the erosion of neocorporatist laborrelations and the promotion of market regula-tion have eroded the strength of labor market‘‘insiders,’’ notably unionized (male) workers instable full-time jobs.4That has weakened thedefence of employment security regulations andcustomary practices preserving job security.Governments in all types of economy havemade it easier to dismiss workers and to‘‘downsize.’’ In doing so they have also made iteasier for firms to alter job boundaries, reduc-ing the rights of existing workers and encour-aging resort to external labor markets, enablingemployers to substitute lower-cost labor for‘‘core’’ workers.

Third, the income security of the employedhas been reduced in many countries, in part bythe removal or weakening of minimum wagelegislation, or by the non-enforcement of ex-isting laws. Among the consequences has beena growth of very low-wage employment, in-cluding jobs paying ‘‘individual’’ rather than‘‘family’’ wages. This has encouraged a substi-tution of women for men and induced laborforce entry by women.

Fourth, in low-income countries in particu-lar, the emphasis put on trade liberalizationand export-led industrialization has had im-plications for women’s economic activity. Thedirect e?ect has been documented by severalanalyses.5Indeed, all countries that have suc-cessfully industrialized have done so only bymobilizing large numbers of (low-paid) womenworkers. Indirectly, the industrialization strat-egy has meant that subsidies for domestic‘‘non-tradables’’ have been cut, often includingstaplefooditemstypicallyproducedbywomen, and structural adjustment programeshave involved deflationary stabilization plansthat, in reducing domestic consumption so asto shift resources to export industries, havealso had adverse e?ects on the living standardsof women producing basic consumer goods.But, as noted earlier, through an increasingemphasisoncost-cuttingcompetitiveness,globalization has also meant a search for waysof lowering labor costs, meaning that firmshave put a greater premium on workers pre-pared or forced to take low-wage jobs. In in-dustrialized and industrializing countries, firmshave turned to forms of labor o?ering theprospect of minimizing fixed non-wage costs.As a result, they have turned increasingly tocasuallabor,contractlabor,outsourcing,home-working and other forms of subcon-tracting.

As part of this flexibilization, there has beenan ‘‘informalization’’ of employment across theworld. Although the dichotomy of ‘‘formal’’and ‘‘informal’’ sectors has always been mis-leading, a growing proportion of jobs possesswhat may be called informal characteristics,i.e., without regular wages, benefits, employ-ment protection, and so on. Such forms ofemployment have been compatible with char-acteristics presumed to be associated withwomen workers — irregular labor force par-ticipation, willingness to work for low wages,static jobs requiring no accumulation of tech-nical skills and status, etc. The informalizationcould thus be expected to be a major factorstimulating the growth of female employmentacross the world.

Fifth, in this process enterprises around theworld have been introducing production tech-niques that have been changing skill and jobstructures in particular ways. Whether therehas been ‘‘deskilling’’ or ‘‘upgrading’’ overall,two trends seem widespread. First, there hasbeen a decline in the proportion of jobs re-quiring ‘‘craft’’ skills learned through appren-ticeship or prolonged on-the-job learning. Suchcrafts have traditionally been mainly the do-main of men, so that their decline and thechanging character of ‘‘skill’’ are likely to haveinfluenced the gender division of labor. Second,there has been a trend to skill ‘‘polarization,’’with a minority of workers required to possessspecialist skills and a majority required topossess minor training, typically impartedthrough ‘‘modules of employable skill,’’ inwhich docility, application, rote learning andrelated ‘‘capacities’’ figure prominently.

This polarization places greater reliance onexternal rather than internal labor markets,since fewer workers are in ‘‘progressive’’ jobswhile more are in ‘‘static’’ jobs involving littleupward mobility or returns to on-the-job con-tinuity. This has weakened one reason for dis-crimination against women, that (whether trueor not) women have a higher labor turnover. Ifthere were less benefit to enterprises fromworkers’ on-the-job experience, that reason fordiscrimination would be removed. Indeed, formany monotonous jobs high labor turnovermay have a positive value for employers, sincemaximum e?ciency may be reached after onlya few months, thereafter plateauing or declin-ing.6

This diminishing return to on-the-job conti-nuity has been one reason for resorting to ca-sual or temporary labor, or for job-rotating,and has been a determinant of the tendency tocollapse job classifications into more broadly based job clusters, such that workers can beshifted from one set of tasks to another fromtime to time. This has been a trend in manylabor markets, and has represented a growth ofjob insecurity that has accompanied the growthof income and employment insecurity markingthe shift to more flexible labor markets.

So, the primary hypothesis is that the grow-ing labor market flexibility and the diverseforms of insecurity have encouraged greaterfemale labor force participation and employ-ment. The evidence presented in the earlierpaper seemed to support this hypothesis. Thequestion is whether the trends continued in thesucceeding decade.

3. GLOBAL FEMINIZATION?

Let us start by considering the changinglevels of female participation in o?cially rec-ognized labor force activities. There has been along debate on the gender bias in o?cial sta-tistics and concepts of labor force participation.The recorded rates of participation have beenseriously a?ected by conceptual and statisticalpractices that have made much of women’swork ‘‘invisible’’ and undervalued. Besidesthese issues (which should always be borne inmind), female labor force participation is de-termined by a mix of economic, demographic,cultural and labor market factors.

It might be useful to reiterate a few stylizedinterpretations:

(a) As initially shown by Ester Boserup, wom-en’s participation in predominantly rural econo-mies has been linked to the type of agriculture,and as a result urbanization and industrializa-tion have not always been associated with a riseor a fall in the rate of participation.8

(b) Marriage, child-bearing and child-raisinghave been barriers to labor force entry and toretention of employment, and these ‘‘barriers’’have been linked to the availability of wage em-ployment, the costs for women of labor forceentry, the type of employment and type of socialtransfers available.

(c) Cultural determinants of participation havebeen widely cited, notably religion and patriar-chal ideology, and these too have been strongerin economies in which work away from thehome has predominated. As with the fertilitydeterminant, most analysts would now be moreskeptical than used to be the case of the strengthof cultural barriers to women’s economic activ-ity, since modifications in working patternsseem remarkably rapid in the face of alterationsin incentives, economic needs and opportuni-ties.

(d) With industrialization based on textiles, gar-ments, electronics and other ‘‘light’’ industry,female participation and employment has tend-ed to rise very sharply.

(e) Traditionally, with growing or high unem-ployment, discriminatory barriers and discour-agement have probably had a greater negativee?ect on women’s labor force participation,and this has tended to dominate the ‘‘additionalworker’’ e?ect of recessions.

According to most analyses, in recent yearsthe negative determinants have been weakenedand the positive factors have been strength-ened. Among the changes have been rising di-vorce rates, declining fertility rates and thepassage in many countries of anti-discrimina-tion legislation. The main factor, however hasbeen the changing nature of the labor market.The concept of regular, full-time wage labor asthe growing type of employment has been giv-ing way to a more diverse pattern, character-ized by ‘‘informalization’’ of employment,through more outworking, contract labor, ca-sual labor, part-time labor, homework andother forms of labor unprotected by laborregulations. Whereas traditionally informaleconomic activities were mainly the means ofsurvival by the rural and urban poor, in recentyears in both industrialized and industrializingcountries there has been a trend in which evenlarger-scale enterprises have been informalizingtheir labor process.

In that context, having done so for the early1970s and 1980s for the first article, we have assembled national-level o?cial data for asmany countries as possible having informationfrom the early 1980s and from the 1990s.9Of-ten the concepts used and the measurementvary, and one should be wary about makingdetailed comparisons. At best, one can paint animpressionistic picture. Fortunately, the trendsdo seem strong enough for us to have reason-able confidence in their validity.

For the past 30 years or so, the trend acrossthe world has been for female labor force par-ticipation to rise, while the male participationrate has been falling.10Tables 1 and 2 showthat in 51% of so-called developing countrieswith available data female labor force partici-pation rose while the male participation ratefell, and in no less than 74% of those countriesthe female rate rose, while in 66% of countriesthe male rate fell. In industrialized countries thedivergence was even greater, with male partic-ipation falling in 95% of all cases. Withincountries, di?erences between male and femalelabor force participation rates have shrunkconsiderably. But at least between industrial-ized countries, there has been no convergencebetween activity rates for women. Within theEuropean Union, for example, the substantialdi?erences between countries have been virtu-ally unchanged over recent years.11In countriesof Central and Eastern Europe, despite theupheavals and economic decline, the levels offemale participation have remained high, al-though they have dropped as if converging tothe (rising) levels of Western Europe. Mostsignificantly, in much of Eastern Europe andthe former Soviet Union male labor force par-ticipation rates have dropped to a greater ex-tent so that the female share of the labor forcehas risen.

A trend brought out by Tables 1 and 2 is thatin a majority of countries in which male par-ticipation fell, total labor force participationrose, suggesting a strong change in the genderdivision of labor and suggesting that female la-bor force entry was more than substituting formen.12With many more women continuously inthelaborforceorfindingiteasiertomoveinandout of it, or combining labor force and otherwork, more women are remaining in the laborforce until a later age. Another interesting pointis that the net increase in overall participationseemed to be greater than over the previousdecade even though the drop in male participationwas much stronger in the later period.13

Among other points emanating from Ta-bles 1 and 2 is that, as observed in the earlierdata summarized in the 1989 article, in thosecountries that have pursued an export-led in-dustrialization strategy as part of a structuraladjustment program, the female labor forceparticipation has been high and has risen. Thisleads to a second series of considerations.

Table 3 shows that in all three regions ofdeveloping countries there has been a tendencyfor the female share of non-agricultural em-ployment to rise, even though women stillcomprise a minority of such employment. Ta-ble 4 gives the patchy time-series data that existon the female shares of manufacturing wageemployment in industrializing countries. Al-though there are relatively few countries withsuch data, they do suggest that the trend hasbeen upward, even though the slight slippage inthe Republic of Korea and Hong Kong mayhave something to do with the changing char-acter of industrial growth in such countries.

In Western Europe and in other industrial-ized countries the female share of non-agricul-tural employment has risen everywhere (exceptin Denmark, where the level has long beenhigh). In Eastern Europe and the former SovietUnion, despite claims to the contrary, women’sshare of industrial employment has remainedvery high and at least in the largest two (theRussian Federation and Ukraine), as well as inSlovakia and Slovenia, women’s relative em-ployment position actually improved after1990, largely because the sectors and jobs heldby men shrunk even more than other sectors.14Table 5 shows a rather more mixed picture oftrends in women’s share of production workersin industry, although definitional di?erencesbecome even greater with data on what is asmaller category of workers, making suchcomparisons hazardous, as exemplified by thecase of Botswana. There is also doubt aboutwhether all countries include unpaid and own-account workers in this category. With thesecaveats, the main finding is that the femaleshare of such jobs remains low. It seems that inlow-income countries barriers to formal formsof wage labor have remained strong, eventhough it is possible that women have not beenseeking such jobs. In rapidly industrializingcountries their share has been higher, as inThailand and Malaysia.Table 5 is consistent with the hypothesis thatit is the spread of more flexible and informalemployment that accounts for much of theupward trend in the female share of the laborforce. One possible reason for the impliedsubstitution of women for men is the lower wage earnings received by women. 

Interna-tional comparisons of wage data are probablyeven more problematical than for other aspectsof labor force participation. But a basic hy-pothesis here is that the increasing globaliza-tion and the more systematic pursuit ofinternational competitiveness have made wageand labor costs more important in determiningthe geographical changes in production andemployment and thus in determining whichgroups are employed.In this regard, one should be wary about in-terpretations of the available data on wage dif-ferentials. Women’s wages may be lower thanmen’s because of job or training discrimination,because of occupational segregation, because ofdirect wage discrimination or because womenare prepared to labor for less, having lower‘‘aspiration wages.’’ The erosion of minimumwage legislation — and the institutional ma-chineryneededtomakeitmeaningful—coupledwiththesanctioningunderstructuraladjustmentprograms of real wage cuts, are likely to inducesubstitution of women for men, partly becausemen are less willing to work for sub-family wagerates and partly because they would be expectedto respond to lower wages by a lower ‘‘e?ortbargain.’’ So, employers would be inclined tohire women more readily. While the promotionof female employment is desirable, this is surelynot the way to achieve it.

The national statistical evidence on wage differentials in developing countries is againdeplorably patchy, and one would be foolhardyto state from what is available that there hasbeen an international trend one way or theother.15

As shown in Table 6, however, in mostcountries where there are time-series data thegender wage earnings’ di?erential has remainedsubstantial.

Intriguingly, it is in rapidly industrializingcountries in which the female share of em-ployment has risen most that show greaterwage di?erentials, although recently it mayhave narrowed marginally in the Republic ofKorea. Although data di?erences and defi-ciencies might explain some of the observedpattern, one hypothesis is that if women’s rel-ative wages are typically lower in SoutheastAsia than in other developing regions of theworld the di?erential may have been both aprimary factor in the rapid industrialization ofthat region and have been perpetuated in partby the character of that industrialization.16

Although this is not the place to try to doc-ument such interpretations, one hypothesis toexplain the combination of wide gender di?er-entials, low wages and rapid industrial growthin the region is that the average social wage islower in much of Southeast Asia than else-where. By this is meant that the individualmoney wage needed to meet a socially accept-able subsistence is lower, both in absolute termsand as a proportion of any person’s social in-come. If one conceives the wage as one part(perhaps none, perhaps all) of a person’s in-come, with other parts coming from family orcommunity transfers, state transfers and enter-prise benefits, the contribution of the variouscomponents is likely to vary according to thetype of economy and society. In SoutheastAsian economies, a woman wage worker hasbeen typically young, single and highly ex-ploitable in part because to a certain extent herwage labor income has been supplemented bytransfers from her (village) community, both atthe time of her wage labor and subsequentlywhen returns to their village.17

Although thispractice exists everywhere, it may be much moresystematic in these countries.18One should alsorecall that women in Southeast Asian factories(as in many other parts of the world) havetypically worked very long work weeks, little orno di?erent from those worked by men.

As for industrialized and Eastern Europeancountries, in the latter gender-based wage dif-ferentials may have been growing, while inWestern Europe they have probably shrunk inrecent years, although in some countries theymay have widened after many years of im-provement.19There and elsewhere, a researchissue is whether the feminization of employ-ment and the lower wages received by womenhave contributed to the growth of income in-equality that has occurred in many parts of theworld.Another facet of the flexibilization of labormarkets has been the rolling back of the public sector, most notably as part of structural ad-justment programes, as part of ‘‘privatization’’initiatives and as a result of the growing prac-tice of outsourcing public service functions. Inmany parts of the world, the public sector hadbeen a leading source of employment growth inthe 1960s, 1970s and early 1980s, and in manycountries (particularly industrialized and East-ern European countries) women have com-prised a higher share of total public sectoremployment than private sector employment.The reversal or slow-down of public sectorgrowth in the late 1980s and 1990s would initself have been expected to lower the femaleshare of total employment, if the average ratioshad remained the same as in the earlier era. Asfar as developing countries are concerned,however, although our data base is not veryrepresentative, Table 7 suggests that the femaleshare of public sector employment has tendedto rise. This continued the trend observed forthe 1980s.20Almost definitionally, a key feature of labormarket flexibilization has been a relative andabsolute growth of non-regular and non-wageforms of employment. There is considerablenational, anecdotal and sub-national data totestify to this trend.21

Most statistical o?ces,however have either not collected informationon casual and other forms of non-regular wagelabor or have started only recently. As far asso-called ‘self-employment’ is concerned (whichmight be taken as a proxy indicator), it is notsurprising that internationallycomparativedata for developing countries are poor.22Theavailable statistics mostly suggest an upwardtrend, notably in Latin America and the Ca-ribbean, as indicated in Table 8.In industrialized economies, numerous sta-tistics show that the relative and absolutegrowth of temporary, casual, contract andpart-time labor have been widespread, sub-stantial and sustained over the past two de-cades, so that in some countries, such as Spain,a majority of all jobs are non-regular.23Some of the most rapidly growing forms offlexible labor are increasing feminization inboth senses of the term — they are absorbingmore women than men and involve less secureworking conditions. An example is teleworking.Many women have been employed in this way, and have been doing so far more systematicallythan men. 

Several empirical studies have shownthat men who have done this sort of work havebeen working at home only for part of the time,whereas women have usually been working full-time at home and in an informal way, leavingthem with only ‘‘second-class citizenship.’’24Relocation of flexible forms of labor havealso been linked not just to feminization but toa certain form of subordinated flexibility. Forinstance, a study of suburbanization of em-ployment in the United States reported that thedecision to relocate was linked to the employ-ers’ preference for married, white women asclerical workers.25Teleworking promises to beone of the fastest growing forms of employmentand a major source of female employment inthe next decade.Finally, in both the 1980s and 1990s women’srate of unemployment relative to men’s fell in aconsiderable majority of industrialized and de-veloping countries. 

Table 9 shows that in 83%of the developing countries covered women’srelative unemployment fell, and that this was the case in 73% of industrialized countries. The table also shows that in a substantial majorityof industrialized countries the female unem-ployment rate was less than the male equiva-lent. Although comparable statistics were notgiven in the earlier article, a longer version ofthat paper did present this information. 

Acomparison those figures with Table 10 indi-cates that in the 1970s female unemploymentrelative to male rose in a majority of industri-alized countries (78%) and in the same numberof developing countries as it fell (45%), whereasin the 1980s it fell in 63% of industrialized countries and 59% of developing countries. Inother words, the drop in women’s relative un-employment has been a fairly recent phenom-enon but the deterioration of men’s positionhas been accelerating.

Given that women’s labor force participationrates and employment have been rising andwomen’s unemployment rates have fallen, thischange cannot be explained by reference towithdrawal from the labor force in times ofhigh overall unemployment. It seems to reflecta considerable erosion in the position of men inlabor markets throughout the world (see Ta-bles 11 and 12). For an example, consider thecase of the United Kingdom. Between the early1960s and the mid-1990s, UK male employ-ment dropped by about four million (excludinga rise in self-employment), so that by the mid-1990s about 4.2 million men aged 16–64 wereeither unemployed or inactive. Meanwhile, fe-male employment rose by nearly three million.These developments could not be explained bydemographic factors.

The developments inother countries might not have been so dra-matic, but throughout the industrialized worldthe trend seems to have continued. For in-stance, in Belgium in 1990 the female–male unemployment ratio was 2.5; by 1993, it was1.8; in Germany in the same period, the ratiofell from 1.5 to 1.1; in Canada, it fell from 1.0to 0.9. 

Women can no longer be regarded as theprimary ‘‘labor reserve,’’ since their labor forceparticipation rates have risen, their share ofemployment has risen and their relative unem-ployment rates have fallen. The rising relativeand absolute levels of male unemployment arecreating a crisis for social and labor marketpolicy, since the welfare state was based on thepresumption of the full employment of men inregular full-time jobs.

This long-term trend should not be inter-preted as implying that women’s position is‘‘good.’’ The reality is that men’s position hasbecome more like that of women, and this isespecially the case with respect to entitlement toa modicum of income security once unem-ployed. 

Even in the European Union, known tohave the most developed of state transfermechanisms, by the mid-1990s only about one-third of all unemployed were receiving unem-ployment benefits, and in this regard (as in somany others) women’s position remains worsethan men’s.26But men’s position has deterio-rated. What should be most worrying is thatthe income security for the unemployed hasbeen declining, due in part to the chroniccharacter of mass unemployment and in part tothe explicit and implicit disentitlement to ben-efits.

4. CONCLUDING REMARKS

There are some who take exception to thenotion of ‘‘feminization’’ of the labor market.The three trends identified in the earlier articlehave however remained powerful and have possibly accelerated over the past decade or so.The types of employment and labor force in-volvement traditionally associated with women— insecure, low-paid, irregular, etc. — havebeen spreading relative to the type of employ-ment traditionally associated with men — reg-ular, unionized, stable, manual or craft-based,etc. In addition, women have been entering, re-entering and remaining in the labor force to agrowing extent. A third trend is that more menhave been forced into the margins of the labormarket, if not out of it altogether. In e?ect,there has been a convergence of male and fe-male patterns of labor force participation.While there has been an overall trend towardmore flexible, informal forms of labor, women’ssituation has probably become less informal,while men’s has become more so.A welcome development is that, according toa recent exhaustive assessment, there has beensome decline in the extent of sex-based occu-pational segregation in most parts of theworld.27However, this too may largely reflectthe weakening position of men rather than anydramatic improvement in the occupationalopportunities of women.

The trends of flexibility and feminizationcombine to pose an historical challenge to so-cial and labor market policy. It is not possibleto presume (as too often has been the case) thatthe ‘‘family wage,’’ ‘‘breadwinner’’ model oflabor force behavior is anything like the norm,either currently or likely to arise in the nearfuture. Social insurance predicated on regular,stable full-time wage labor with ‘‘temporaryinterruptions in earning power’’ does not pro-vide women, or increasingly men, with socialprotection. Means-tested ‘‘social safety nets’’do not do so either.

So, among the challenges are the need to re-form systems of social protection. There is acorresponding need to promote alternativeforms of collective institution to protect andenhance the status of vulnerable groups in labormarkets, and a need to combine flexibility withsteadily improving economic security. Women’sgrowing involvement in labor force activities isto be welcomed as facilitating a trend towardgender equality, and should be strengthened.But the conditions in which women and men aretypically in the labor market do not seem tohave been improving. The trend is towardgreater insecurity and inequality. Reversing thattrend, which is associated with labor flexibility,is the most important labor market and socialpolicy challenge of all.

NOTES

1.Construction laborer in India. van der Loop (1996),p. 390. There was no suggestion that either the laborer orthe author appreciated the irony of the statement.2.Standing (1989), pp. 1077–1095.3.Some claim that globalization has not had muche?ect on labor markets (IMF April). But, not only isthere evidence of export-oriented industrialization inmany developing countries as well as ‘‘de-industrializa-tion’’ in industrialized countries, but there is evidence of‘‘whipsaw bargaining’’ by managements, along the linesthat unless workers accept lower wages and lessemployment security, etc., the firm would relocate orchannel new investment elsewhere.4.For a perspective on issues raised in this paragraph,see Standing (1997).5.See, for instance, Wood (1991). Wood concludedthat ‘‘developing countries which exported a risingproportion of their manufacturing output to the north(sic) tended to employ a rising proportion of females intheir manufacturing sectors’’ (p. 171).6.

In visits to electronics factories in Malaysia in the1980s, it was interesting to find that many managersexpected and even wanted the young women workers toremainintheirjobsforonlytwoorthreeyears.Theyknewthatafterashorttimetheirphysicalproductivitydeclined,often as a result of illness or spinal or optical injuries.7.The determinants of female work patterns aroundthe world were synthesized some time ago (Standing,1978, 1981). There are strong grounds for radicallyoverhauling conventional labor force statistics.8.Boserup (1970).9.Of course, there has been a vast amount of otherdata and analysis in recent years. This is an attempt tosee what one can tell from o?cial national data.10.For trends in the post-1945 era, see Standing(1981), chapter 1. For more recent years, for industri-alized countries, see, for example, Meulders et al. (1993).11.Rubery et al. (1995), p. 5.12.One cannot say anything conclusive from such dataabout the overall rate of labor absorption, let alone theelasticity of employment with respect to growth, etc.,simply because the quantity of labor and rate ofemployment relative to unemployment are not capturedby participation rate data.13.Standing (1989), Tables 1 and 2, pp. 1081–1082.14.This has been documented through the RussianLabor Flexibility Survey and the Ukrainian equivalent,which have been the biggest surveys of industrialenterprises conducted in those countries, covering hun-dreds of thousands of workers. See, for instance,Standing (1996a, b); Standing and Zsoldos (1995).15.Case studies have suggested that in the 1970s and1980s gender-based wage di?erentials did decline. Ankerand Hein (1986).16.Some analysts have asked why, in the light of thelarge wage di?erentials, did the workforces not becomeentirely female. Feminist (and Marxisant) interpretat-ions would postulate that disciplinary and other divide-and-rule tactics would induce a mixed-gender strategy inthe workplace. See, for instance, Elson and Pearson(1981): Joekes (1993), p. 1717.This does not rule out cash remittances fromworkers in urban areas, but the two-way process haslong characterized labor circulation in the region. Toooften studies of remittances only consider the flow fromurban to rural areas.18.One could argue that families are exploitedthrough the superexploitation of young women workers,meaning that they could scarcely survive on their wageearnings alone. Another way of putting the argument ofthe text is that the wage is less than the cost ofreproducing their labor power in part because the youngwoman’s family has a rural production base or isworking in periurban informal economic activities. Insouthern Africa, by contrast, the urban wage has beenhigher in part because that has been needed to secure astable and productive supply of wage labor becauseurban-industrial workers have rarely received much inthe form of community transfers and have been expectedto support rural households through remittances.19.Gonzalez (1995).20.Standing (1989), p. 1087. 

The trend suggests ateasing question: Since in some countries the femaleshare of the public sector has been lower than in theprivate economy, would a cut in the public sector haveboosted the female share of total employment? Wouldthe reverse have applied in countries where the femalepublic sector share has been higher than in the privateeconomy?21.See, for instance, Bettio et al. (1996).22.There is good reason to believe that commonly andindefensibly women working for an income other than awage are classified as ‘‘unpaid family worker’’ whereas aman in similar work would be classified as ‘‘self-employed’’ or ‘‘own account.’’ This is one of the manyreasons for not collecting labor force participation dataon the basis of a priori categories if possible.23.Standing (1997), Table 3.24.Lie (1985).25.Moss (1984).26.Meulders (1996), p. 24. This remarkable fact isdocumented in more detail in Standing (forthcoming).27.Anker (1998).

REFERENCES

Ankerf, R., Gender and Jobs: Sex Segregation ofOccupations in the World (Geneva: ILO; 1998).Anker, R., and C. Hein, Sex Inequalities in UrbanEmployment in the Third World (Basingstoke: Mac-millan, 1986).Bettio, F., Rubery, J., and M. Smith, ‘‘Gender, flexibilityand new employment relations,’’ Mimeo (Manches-ter: University of Sienna and UMIST, 1996).Boserup, E., Women’s Role in Economic Development(New York: St.Martin’s Press, 1970).Elson, D., and R. Pearson, ‘‘The subordination ofwomen and the internationalisation of factory pro-duction and of marriage and the market,’’ in K.Young, C. Wolkowitz, and O.R. McCullogh (Eds.),Of Marriage and the Market: Women’s subordinationin International Perspective (London: CSE Books,1981).Gonzalez, P., ‘‘Indicators of the relative performance ofwomen in the labour market,’’ Mimeo (Cambridge:University of Cambridge, 1995).International Labor Organization, Year Book of LabourStatistics (Geneva: ILO, various years).International Monetary Fund, World Economic Outlook(Washington, DC: IMF, 1997).Joekes, S., ‘‘The influence of international trade expan-sion on women’s work,’’ Paper prepared for ILOProject on Equality for Women in Employment(Geneva: ILO, 1993).Lie, M., ‘‘Is remote work the way to’’ ‘‘the Good Life’’‘‘for women as well as men?’’ (Trondheim, Norway:Institute for Industrial Research, 1985).Meulders, D., ‘‘Women and the Five Essen Priorities’’(Brussels: Universite Libre de Bruxelles, 1996).Meulders, D., R. Plasman, and V. Vander Stricht,Position of Women on the Labour Market in theEuropean Community (Aldershot: Dartmouth Pub-lishing, 1993).Moss, M.L., New Telecommunications, Technologies andRegional Development (New York: New York Uni-versity, 1984).Rubery, J., M. Smith and C. Fagan, Changing Patternsof Work and Working Time in the European Unionand the Impact on Gender Divisions. Report for theEqual Opportunities Unit, European CommissionDirectorate General V, V/6203/95-EN (Brussels:European Directorate General, April 1995).Standing, G., Shadow of the Future: Global Flexibilityand Distributive Justice (Harmondsworth: Macmil-lan, forthcoming).Standing, G. ‘‘Globalisation, labour flexibility andinsecurity: The era of market regulation,’’ EuropeanJournal of Industrial Relations, Vol.3, No.1 (March1997), pp. 7–38.Standing, G., Russian Unemployment and EnterpriseRestructuring: Reviving Dead Souls (Basingstoke:Macmillan, 1996a).Standing, G., ‘‘The ‘‘shake out’’ in Russian factories:The RLFS fifth round, 1995’’ (Geneva: ILO, 1996b).Standing, G., and L. Zsoldos, ‘‘Labour market crisis inUkrainian industry: The 1995 ULFS’’ (Geneva: ILO,1995).Standing, G., ‘‘Global feminisation through flexiblelabour,’’ World Development, Vol. 17, No.7 (1989).Standing, G., Labour Force Participation and Develop-ment (Geneva: ILO, 1978).Standing, G., Labour Force Participation and Develop-ment (Geneva: ILO, 1981).Van der Loop, Theo, Industrial Dynamics and Fragment-ed Labour Markets: Construction Firms and Labour-ers in India (New Delhi: Sage Publications, 1996).Wood, A., ‘‘North-South trade and female labourin manufacturing: An asymmetry,’’ Journal ofDevelopmentStudies,Vol.27,No.2(January1991).